Reliable and valid estimates of the prevalence of substance abuse are crucial to effective policy development, prevention, and evaluation in schools and in the workplace. Numerous methods for deriving these estimates have been tried, but all have significant flaws: underreporting, obtrusiveness, expense, etc. Prior research suggests that the item count/ paired list (IC/PL) method can be adapted to provide accurate prevalence estimates, unobtrusively and at low cost. Long-term objectives are to develop IC/PL questionnaires for use in schools or the workplace. Reliability and validity of estimates will be established through a series of field tests; criterion validity will include comparisons with direct self-report and with physiological tests for substance use. The project represents the first time IC/PL has been adapted for this purpose and, if successful, a major breakthrough in survey technology for substance abuse assessment. School and workplace surveys, and social science research represent the major commercial applications.